Subdomain vs Main Domain Penalties

[Prospector-Plastics]

We have a client who's main root.com domain is currently penalized by Google, but the subdomain.root.com is appearing very well. We're stumped - any ideas why?

[jesse-landry]

hmmm...

do they point to the same IP address?

[AdamThompson]

This isn't necessarily surprising. Penalties and negative ranking algorithms can be applied at a page level, a subdomain level, a root domain level, etc.

For example, HubPages used subdomains to help escape from a Panda slap.

Another example: Google placed a manual penalty on a single page of BBC's website.

[ScottJames13]

From my understanding subdomains are considered completely separate from root domains unless you have a 301 redirect or conical that tells search engines you want them to consider the root or the subdomain to be the same; for example, http://www.yourdomain.com (subdomain) points to http://yourdomain.com

Therefore, you could have a subdomain out rank a root domain, or in your case a root domain penalized and the subdomain continue to rank well. The fact that they share an IP address shouldn't affect all the domains under that IP as many websites are on shared hosting which use the same IP address.

[Prospector-Plastics]

Thank you all for your insight - good stuff, but still stumped.

Here's everything we know that might help point out why the main domain (ie www.root.com) was penalized by Google. We redirect root.com to www.root.com with a 301 redirect, and it is setup this way in Google Webmaster Tools too.

December 9, 2011 - the site's 404 error page was incorrectly setup as a 200, resulting in a quick bloat of 1 million plus pages. The website dropped from Google immediately. The error page was correctly setup 2 days later. The site still appeared in Google's index via site: query. However the site didn't reappear in Google's SERPs until May 2, 2012.

October 25, 2012 - the website again drops from Google for an unknown reason. We then moved a significant portion of content from www.root.com to subdomain.root.com. Pages from subdomain.root.com began appearing within 3 days as high they appeared previously on Google. From December 9, 2011 throughout this entire time we were correcting any errors reported in Google Webmaster Tools on a daily basis.

February 26, 2013 - The website yet again is dropped from Google, the subdomain.root.com continues to appear and rank well.

Due to moving most of the content from www.root.com to subdomain.root.com, the index for www.root.com from October 2012 dropped from 142,000 slowly to an average of 21,400 ending at today's 4,230. However this index count fluctuates greatly every few days (probably due to moving content from www.root.com to subdomain.root.com).

Of note, the site is NOT a content farm, but legitimate unique technical content that is hosted for hundreds of clients.

Again any ideas are most welcome!

[Maximise]

Subdomains and root domains are not necessarily always owned by the same person and therefore will not always be given the same penalties. As Scott mentioned, they are seen as different sites.

e.g. If I create a new WordPress account and create me.wordpress.com and then build a black hat site which gets penalized, this is not going to affect you.wordpress.com or  www.wordpress.com.

[MarieHaynes]

I think the answer here depends on whether or not you have actually been penalized and why the site is dropping out of the SERPS.  Do you have a warning in WMT?  If not, then you're probably not penalized.

It's unlikely to be Penguin because Penguin did not refresh lately.  Similarly, Panda did not refresh on the days you mentioned.  So, it's not likely a penalty but rather some type of site structure issue.

Is there duplicate content between the subdomain and the root?  If so, then Google will choose one as the owner and not show the other prominently.  Any issues with robots.txt?  Are the canonicals set correctly?  Any chance of accidental noindexing?

[Prospector-Plastics]

Thank you Marie,

We 301 redirect any traffic going to root.com to www.root.com, and any content that we moved from www.root.com to subdomain.root.com has been completely removed from www.root.com. There doesn't appear to be any duplicate content between the two. There is some duplicate content that we treat with canonicals on subdomain.root.com - very small portion of total pages (less than 1%).

As for your other questions, no warnings in WMT. Robots txt file looks clean, canonicals are in place correctly, and no accidental non-indexing that we know of.

Here is the actual site that might help to look at:

http://www.ides.com
http://plastics.ides.com/materials
http://www.ides.com/robots.txt

[MarieHaynes]

You have some significant duplicate content issues with www.ides.com.  There is not a lot of text on your home page and what is there is duplicated in many places across the web.

Your second level pages are all just links.  I would noindex, follow these.

I looked at two inner pages:

http://plastics.ides.com/generics/6/alphamethylstyrene-ams - extremely thin content

Here is a Google search for text I copied from the styrene-acrylonitrile page.  There are 247 pages that use this opening sentence.

My guess is that this is indeed a Panda issue.  But please know that I've only just taken a quick look so I can't say for sure.  What doesn't make sense is that your traffic drops don't happen on Panda dates which really should be the case if it was Panda.

Panda definitely can affect just one part of a site (such as a root and not a subdomain).  I would work on making these pages completely unique and also noindexing the thin pages.

[Prospector-Plastics]

Extremely helpful insight Marie - I will be contacting you directly soon.

It appears that the duplicate content you've found (and other dupe content we've found) is actually our content that other sites have repurposed. Seems like Google has determined our site as the culprit, so this would be an issue we need to address - the only thought that comes to mind right away is adding an 'Author' tag, then start working on what appears to be a hefty cleanup project, something that looks like you are an expert on and will most likely be working directly with you in the near future!

The 2nd level pages that have little content and lots of links are 'noindex,follow' but I'm nervous about the number of these tags throughout our site which could be seen as spammy to a search engine. Of note, the 2nd level page section you have found ranks quite well since it is a subdomain which is interesting. Our suspicion is that since we made the 404 (200 success) error that Google detected on Dec. 9, 2011, we have been on some sort of Google 'watch-list' and any little thing we do incorrectly that they find, we immediately are penalized.

The homepage description of our company is reused on industry directories that we are listed on, so perhaps we must consider re-writing our description to be unique, and adding more content to the homepage would be a good thing and is certainly easily doable.